AI-based staple map automatic generation method and system

Abstract

The invention discloses an AI-based staple map automatic generation method and system, and particularly relates to the staple map generation technology field, a radio frequency transmit-receive device and an unmanned aerial vehicle are used to collect boundary line inflection point coordinates of a staple and intersection point coordinates of the staple and an adjacent staple boundary line, and the inflection point coordinates are connected in sequence to obtain a staple boundary line; then, a land parcel boundary line and a land parcel area are corrected and judged, a land parcel area correction link is directly entered if a land parcel boundary line correction task is not triggered, a land parcel boundary point correction link is entered if a land parcel area correction task is not triggered, and a point which can better represent a land parcel trend is screened out from multiple inflection points to serve as a boundary point; according to the method, the problem that the boundary point processing workload is large is solved, meanwhile, the land parcel area is controlled to be consistent with the actual land parcel area, the land parcel boundary line is controlled to be consistent with the actual land parcel boundary line, the four boundaries of the land parcel are analyzed, and the accuracy of a subsequent land parcel map is guaranteed to a certain degree.

Description

Technical Field

[0001] The present invention relates to the technical field of cadastral map generation, and more specifically, the present invention relates to an AI-based automatic cadastral map generation method and system. Background Art

[0002] A cadastral map is a sub-cadastral map that describes the location of a lot, the relationship between boundary points and lines, and the numbers of adjacent lots. When making a cadastral map, it is necessary to accurately locate the boundary points of the lot, the boundary lines of the lot, and the information of the four boundaries of the lot. Unclear lot ownership information will greatly affect the validity of the cadastral map. Therefore, an accurate and efficient method is needed to obtain lot ownership information.

[0003] The existing automatic cadastral map generation methods obtain the ground images of the lot and its surrounding areas by using an unmanned aerial vehicle (UAV) for aerial photography to make a base map, and then carry out the work of household surveys and boundary identification with the base map, and then determine the boundary points of the lot. After connecting the boundary points of the lot, the boundary lines of the lot can be obtained, and then the information of the four boundaries of the lot can be obtained, which can ensure the accuracy of cadastral map production to a certain extent.

[0004] However, there are still some problems with the existing methods: the existing methods need to carry out household surveys and boundary identification. The obtained boundary points of the lot are mostly the inflection points of the lot. For a lot with more inflection points, there are up to thousands of boundary points generated, resulting in a more complicated description of the boundary points and their directions, greatly increasing the workload of determining lot ownership information, and still need to further improve the efficiency of determining lot ownership information. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an AI-based automatic cadastral map generation method and system, which uses a radio frequency transmitter to collect lot ownership information, uses an unmanned aerial vehicle to obtain lot geographical data, and then integrates and corrects the lot ownership information, and screens out the points that can better represent the lot direction from many inflection points as boundary points, solving the problem of large workload in boundary point processing.

[0006] To achieve the above object, the present invention provides the following technical solution: an AI-based automatic cadastral map generation method, including the following steps: S1. Collect lot data: use a radio frequency transmitter to collect lot ownership information, and use an unmanned aerial vehicle to obtain lot geographical data; S2. Integrate and correct lot ownership information: correct the boundary lines, areas, and boundary points of the lot in the collected lot ownership information; S3. Analyze the four boundaries of the lot: use the method of positive rectangular side projection to divide the boundary line into four parts: east, south, west, and north, and obtain the information of the four boundaries of the lot based on the set four-boundary determination rules; S4. AI Recognition and Annotation of Parcel Geographic Data: Recognize and annotate the collected parcel geographic data based on AI algorithms; S5. Create a Parcel Map Template: Use ArcGIS to create a parcel map template; S6. Automatically Generate Parcel Maps in Batches: Run a Python script in the command prompt of ArcMap. After automatically reading the parcel attribute data packet, the script generates a parcel map for each parcel according to the MXD template; S7. Inspection and Revision of Parcel Maps: Check whether the parcel boundaries are accurate, the text annotations are correct, and the map elements are complete. If any one or more of the above are not the case, the parcel map needs to be revised. After the revision is completed, export the parcel map to a preset image format; S8. Data Collection for Parcel Map Generation: Collect data for correcting parcel ownership information, data for inspecting and revising parcel maps, and data for recognizing parcel geographic data; S9. Quality Evaluation of Parcel Map Generation: Process the collected data for correcting parcel ownership information, data for inspecting and revising parcel maps, and data for recognizing parcel geographic data, and then evaluate the quality of parcel map generation.

[0007] To achieve the above objectives, the present invention provides the following technical solution. An AI-based automatic parcel map generation system, implementing the above-mentioned AI-based automatic parcel map generation method, includes: Parcel Data Collection Module: Use a radio frequency transmitter to collect parcel ownership information, and use a drone to obtain parcel geographic data; Parcel Ownership Information Correction Module: Used to correct the parcel boundary lines, parcel areas, and parcel boundary points in the collected parcel ownership information; Parcel Four-Boundaries Analysis Module: Use the method of projecting the sides of a regular rectangle to divide the boundary line into four parts: east, south, west, and north, and obtain the parcel four-boundaries information based on the set four-boundaries determination rules; Parcel Geographic Data Recognition Module: Recognize and annotate the collected parcel geographic data based on AI algorithms; Parcel Map Template Creation Module: Use ArcGIS to create a parcel map template; Parcel Map Automatic Generation Module: Run a Python script in the command prompt of ArcMap. After automatically reading the parcel attribute data packet, the script generates a parcel map for each parcel according to the MXD template; Parcel Map Self-Inspection and Revision Module: Check whether the parcel boundaries are accurate, the text annotations are correct, and the map elements are complete. If any one or more of the above are not the case, the parcel map needs to be revised. After the revision is completed, export the parcel map to a preset image format; Parcel Map Generation Data Collection Module: Used to collect data for correcting parcel ownership information, data for inspecting and revising parcel maps, and data for recognizing parcel geographic data; Parcel Map Generation Quality Evaluation Module: After processing the corrected data of the parcel ownership information, the inspection and correction data of the parcel map, and the identification data of the parcel geographical data collected, it evaluates the quality of the parcel map generation.

[0008] Technical Effects and Advantages of the Present Invention: The present invention issues a certain number of radio frequency transmitters to the parcel owners or the persons in charge of parcel ownership units. After receiving the radio frequency transmitters, the parcel owners or the persons in charge of parcel ownership units input the parcel number, the name of the parcel owner or the parcel ownership unit, the parcel area, and the radio frequency transmitter number information into the radio frequency transmitters, and arrange the radio frequency transmitters at the inflection points of the parcel boundary lines and the intersection points with the boundary lines of adjacent parcels in sequence according to the numbers. Transmit the input information of the radio frequency transmitters as radio frequency communication signals, and control the unmanned aerial vehicle to obtain the orthophoto real - scene image data of the parcel area by using the configured high - definition camera equipment according to the set flight route. During this period, the radio frequency communication signals sent by each radio frequency transmitter are received by the radio frequency communication signal receiving equipment arranged on the unmanned aerial vehicle, so that the parcel number, the name of the parcel owner or the parcel ownership unit, the parcel area, the coordinates of the inflection points of the parcel boundary lines, and the coordinates of the intersection points of the parcel boundary lines and the boundary lines of adjacent parcels can be obtained, avoiding the workload of on - site boundary indication. Then, the parcel boundary lines and the parcel area are corrected and judged. When the parcel boundary line correction task is not triggered, it directly enters the parcel area correction link. When the parcel area correction task is not triggered, it directly enters the parcel boundary point correction link. Among numerous inflection points, the points that can better represent the parcel trend are selected as boundary points, solving the problem of large workload in boundary point processing. At the same time, it controls the parcel area to be consistent with the actual parcel area, the parcel boundary lines to be consistent with the actual parcel boundary lines, and analyzes the four - boundaries of the parcel, ensuring the accuracy of the subsequent parcel map to a certain extent. Description of the Drawings

[0009] Figure 1 It is the method step diagram of the present invention.

[0010] Figure 2 It is the system structure block diagram of the present invention. Detailed Embodiments

[0011] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0012] As Figure 1 shown, this embodiment provides an AI - based automatic parcel map generation method, including the following steps: S1. Collect land parcel data: Use a radio frequency transmitter to collect land parcel ownership information, and use a drone to obtain land parcel geographical data; Furthermore, the land parcel ownership information includes the land parcel number, land parcel boundary line, first boundary point, land parcel area, name of the land parcel owner or land parcel ownership unit. The land parcel geographical data is the orthophoto real scene image data of the land parcel ground. The land parcel boundary line is the ownership boundary line between the land parcel and adjacent land parcels.

[0013] Furthermore, the specific steps for collecting land parcel data are as follows: A1. Distribute a certain number of radio frequency transmitters to the person in charge of the land parcel owner or land parcel ownership unit. The radio frequency transmitter has the functions of transmitting radio frequency communication signals and inputting information; A2. After receiving the radio frequency transmitter, the person in charge of the land parcel owner or land parcel ownership unit inputs the land parcel number, name of the land parcel owner or land parcel ownership unit, land parcel area, and radio frequency transmitter number information into the radio frequency transmitter, and arranges the radio frequency transmitters at the inflection points of the land parcel boundary line and the intersection points with the boundary lines of adjacent land parcels in sequence according to the numbers. The numbers of the radio frequency transmitters arranged at the inflection points of the land parcel boundary line are a1, a2,..., an in sequence, and the numbers of the radio frequency transmitters arranged at the intersection points with the boundary lines of adjacent land parcels are b1, b2,..., bn in sequence. For the radio frequency transmitter that is both at the inflection point of the land parcel boundary line and the intersection point with the boundary line of adjacent land parcels, it has both numbers at the same time; In this embodiment, it should be specifically noted that if the 9th inflection point on the land parcel boundary line is also the 1st intersection point with the boundary line of adjacent land parcels experienced, the number of the radio frequency transmitter arranged at this point is a9b1.

[0014] A3. Transmit the input information of the radio frequency transmitter as a radio frequency communication signal, and control the drone to use the configured high-definition camera equipment to obtain the orthophoto real scene image data of the land parcel area according to the set flight route. During this period, receive the radio frequency communication signals sent by each radio frequency transmitter through the radio frequency communication signal receiving device arranged on the drone, and obtain the land parcel number, name of the land parcel owner or land parcel ownership unit, land parcel area, radio frequency transmitter number information, and geographical location coordinates of the radio frequency transmitter input in the received radio frequency transmitter; Specifically in this embodiment, the drone is equipped with a high-definition camera device and a radio frequency communication signal receiving device. Common high-definition camera devices include high-resolution CCD digital cameras, laser scanners, and multispectral cameras. High-resolution CCD digital cameras have a high resolution and can capture fine textures and details of the ground or target areas. Laser scanners measure distances and generate 3D models by emitting laser beams and receiving the reflected signals. Multispectral cameras can capture spectral information in different bands and can be used to analyze vegetation cover and soil types on the ground surface. In this embodiment, a high-resolution CCD digital camera is selected to be configured on the drone, and high-quality orthophoto real-scene image data of the ground of the land parcel can be obtained, which is beneficial to improving the accuracy of subsequent identification of the components within the land parcel. Common radio frequency communication signal receiving devices include radio frequency receivers and GPS receivers. Radio frequency receivers can ensure the stability of receiving radio frequency communication signals emitted by radio frequency transmitters on the ground of the land parcel. GPS receivers can determine the precise position data of the used radio frequency transmitters. In this embodiment, a radio frequency receiver is selected to be configured on the drone to ensure the stability of receiving signals.

[0015] A4. Connect the corresponding geographical location coordinates in sequence according to the numbers of the radio frequency transmitters arranged at the inflection points of the land parcel boundary lines. The generated closed connection line is the land parcel boundary line. The numbers of the a1a2 section of the boundary line, the a2a3 section of the boundary line,..., the ana1 section of the boundary line are x1, x2,..., xn in sequence. Mark the geographical location coordinate points and numbers of the radio frequency transmitters arranged at the inflection points of the land parcel boundary lines as the third boundary points and the corresponding numbers, and mark the geographical location coordinate points and numbers of the radio frequency transmitters arranged at the intersection points with the boundary lines of adjacent land parcels as the first boundary points and the corresponding numbers.

[0016] S2. Integrate and correct the land parcel ownership information: Correct the land parcel boundary lines, land parcel areas, and land parcel boundary points in the collected land parcel ownership information. Furthermore, the specific steps for integrating and correcting the land parcel ownership information are as follows: B1. Retrieve the land parcel boundary lines and the boundary lines of adjacent land parcels, and observe whether there are other overlapping areas between the corresponding included land parcel areas and the adjacent land parcel areas except for the boundary lines. If there are, trigger the land parcel boundary line correction task; if not, do not trigger the land parcel boundary line correction task and directly enter the land parcel area correction link. Specifically in this embodiment, when triggering the correction task of the land parcel boundary line, the first execution plan is preferentially selected, that is, directly correcting the boundary line by retrieving the land ownership materials of the land parcel and its adjacent parcels. If the boundary line in the land ownership source materials is unclear or the boundary line is inconsistent with the field situation, the second execution plan is selected, that is, notifying the land parcel owner or the person in charge of the land parcel ownership unit and the land parcel owner or the person in charge of the land parcel ownership unit of the adjacent parcel and the surveyors to conduct on-site boundary indication. After the correction of the land parcel boundary line is completed, update the geographical location coordinates and numbers of the first boundary point and the third boundary point, as well as the land parcel boundary line number.

[0017] B2. When entering the land parcel area correction link, check the coordinates and corresponding numbers of each layout point of the radio frequency transmitter to ensure that the coordinates and numbers of the radio frequency transmitter are arranged clockwise or counterclockwise along the boundary line. Then substitute the coordinates of the radio frequency transmitter into the Gauss area formula to calculate the land parcel area Ax. Based on the calculated land parcel area and the received land parcel area, calculate the area deviation coefficient αm. The area deviation coefficient is the ratio of the absolute value of the difference between the calculated land parcel area Ax and the received land parcel area As to the smaller value of the two. The specific formula is: , when the calculated land parcel area Ax is the same as the received land parcel area As, αm = 0. Compare the calculated area deviation coefficient with the preset upper limit value of the area deviation. If the calculated area deviation coefficient is greater than the upper limit value of the area deviation, trigger the land parcel area correction task; otherwise, do not trigger the land parcel area correction task and enter the land parcel boundary point correction link; Specifically in this embodiment, the Gauss area formula is specifically: , after the calculation is completed, use Ax to represent the calculated land parcel area, (xi, yi) is the coordinate of the i-th layout point of the radio frequency transmitter of the land parcel polygon, and n is the number of radio frequency transmitters. Now a set of coordinates of the layout points of the land parcel radio frequency transmitter is given to demonstrate the process of calculating the land parcel area using the Gauss area formula: the coordinates of the layout points of the radio frequency transmitter are pa1(30m, 50m), pa2(30m, 150m), pa3(150m, 240m), pa4(150m, 50m), then the land parcel area is .

[0018] Specifically in this embodiment, when triggering the land parcel area correction task, confirm the positions of the land parcel radio frequency transmitters one by one, correct the positions of the radio frequency transmitters with incorrect positions. If the positions of the radio frequency transmitters do not need to be corrected, directly replace the received land parcel area with the calculated land parcel area.

[0019] Specifically in this embodiment, after the correction of the land parcel boundary line or the area correction or the land parcel area correction, if there are changes in the inflection points of the land parcel boundary line or the intersection points of the land parcel boundary line and the boundary lines of adjacent parcels, update the corresponding boundary point numbers in turn.

[0020] B3. Extract the first boundary points to form a point set. Starting from the first boundary point numbered b1, calculate the straight-line distance D1 between the first boundary point numbered b1 and the first boundary point numbered b2 based on the point coordinates and the Euclidean distance formula. At the same time, obtain the boundary line distance L1 between b1 and b2, calculate L1 / D1, and compare the calculation result with a preset value. If the calculated value is greater than the preset value, it is determined that a new boundary point needs to be inserted between b1 and b2; otherwise, no new boundary point needs to be inserted, and continue to determine whether to insert a new boundary point between the first boundary point numbered b2 and the first boundary point numbered b3. B4. When it is determined that a new boundary point needs to be inserted between b1 and b2, find the point with the farthest distance from the boundary line between b1 and b2 and the straight line connecting b1 and b2 as the new boundary point, insert it between b1 and b2, and mark the new boundary point as the first boundary point with the number recorded as b2. The numbers of the original first boundary points numbered b2, b3,..., bn are automatically shifted back one place and changed to b3, b4,..., bn+1. B5. Repeat steps B3 and B4 until the ratio of the boundary line distance to the straight-line distance between any two adjacent first boundary points is less than or equal to the preset value, then the correction of the point set of the first boundary points is completed.

[0021] S3. Analyze the four boundaries of the land parcel: Use the method of projecting the sides of a regular rectangle to divide the boundary line into four parts: east, south, west, and north, and obtain the four-boundary information of the land parcel based on the set four-boundary determination rules. Furthermore, the specific steps for analyzing the four boundaries of the land parcel are as follows: C1. First, use the method of projecting the sides of a regular rectangle to divide the boundary line into four parts: east, south, west, and north. C2. For all adjacent land parcels of this land parcel, if an adjacent land parcel has a common side with this land parcel in any one of the four directions of east, south, west, and north of this land parcel, then include this adjacent land parcel in the four-boundary information of the corresponding direction of this land parcel. If an adjacent land parcel has common sides with multiple parts of this land parcel at the same time, then include this adjacent land parcel in the four-boundary information of the corresponding multiple directions of this land parcel.

[0022] Specifically in this embodiment, a set of examples are given to explain the process of analyzing the four boundaries of a land parcel: This land parcel is denoted as F1, and the adjacent land parcels are F2, F3, F4, F5, F6, F7, F8, and F9 respectively. Only the adjacent land parcels F4 and F8 have boundary points with this land parcel F1. F2 and F1 have common sides on both the west and north. F3 and F1 have a common side only on the north. F5 and F1 have common sides on both the north and east. F6 and F1 have a common side only on the east. F7 and F1 have common sides on both the south and west. F9 and F1 have a common side only on the south. Therefore, the east boundary of this land parcel F1 is F5 and F6, the west boundary is F2 and F7, the south boundary is F7 and F9, and the north boundary is F2, F3, and F5. For an enclave (isolated land parcel), since it has no adjacent boundary lines, its four boundaries cannot be obtained through the above method. The solution is to determine whether this land parcel is completely contained within another land parcel. If so, the four boundaries of this land parcel are all the other land parcel. That is, if F1 is completely within the area of land parcel F10, then the four boundaries of F1 are all F10.

[0023] S4. AI recognition and annotation of land parcel geographic data: Recognize and annotate the collected land parcel geographic data based on the AI algorithm; Furthermore, the specific operations of AI recognition and annotation of land parcel geographic data are as follows: Recognize the object type elements within the land parcel, highlight the edge outlines of buildings, structures, roads, and topographic features within the land parcel, highlight the boundary lines and boundary points. Mark the buildings within the land parcel with the building name as the data label, mark the structures within the land parcel with the structure name as the data label, mark the topographic features within the land parcel with the topographic feature name as the data label, mark each section of the boundary line with the boundary line number and the boundary line length as the data label, and mark the boundary points with the boundary point number and coordinates.

[0024] S5. Create a land parcel map template: Use ArcGIS to create a land parcel map template; Furthermore, the specific steps for creating the land parcel map template are as follows: D1. Open ArcMap to create a layout, set the size, and then add a title, legend, and north arrow in the layout; D2. Add the land parcel layer, MappingIndex layer, boundary point layer, boundary line layer, annotation layer, building and structure layer, adjacent land parcel layer, and the line vector layer converted from the QSDW land parcel layer, and set the display order, symbols, and styles of each layer; Specifically in this embodiment, it should be noted that the cadastral map layer is the core layer, which is used to display the boundaries, shapes and extents of cadastral parcels. The cadastral map layer is usually created based on the cadastral polygon data in the geodatabase to ensure that each cadastral parcel has a clear boundary and extent definition; the boundary points are the key nodes on the cadastral parcel boundary, and the boundary lines are the line segments connecting these nodes, jointly constituting the boundary of the cadastral parcel. These two layers are used to accurately represent the boundary position of the cadastral parcel to ensure the accuracy of the ownership boundary; the MappingIndex layer serves as the index layer for the automatic mapping code, which is used to traverse each cadastral parcel and extract relevant information from the attribute table; the annotation layer is used to add necessary text annotations, such as cadastral parcel numbers, land use codes, areas, right holders, house numbers, etc. These annotations help users quickly identify and understand the relevant information of the cadastral parcels; the building layer is used to display information such as the location, level and structure of buildings or structures within the cadastral parcel, which helps users understand the layout and nature of the buildings or structures within the cadastral parcel; the adjacent cadastral parcel layer is used to display information of other cadastral parcels adjacent to the current cadastral parcel, including cadastral parcel numbers, boundary lines, etc., which helps users understand the relationship between the current cadastral parcel and the surrounding cadastral parcels; the line vector layer converted from the QSDW cadastral map layer is used to display the boundary of the cadastral map and may need to be cropped and processed through program code to meet specific display requirements. For example, only a small section of the line segment is displayed as a "small thorn".

[0025] D3. Use ArcPy to write an automatic mapping script that includes operations such as reading layer attributes, drawing map elements, updating text elements, intelligently adjusting the map display scale, and intelligently placing annotations; D4. Save it as an MXD template file.

[0026] S6. Automatically generate cadastral maps in batches: Run the Python script in the command prompt of ArcMap. After automatically reading the cadastral parcel attribute data packet, the script generates cadastral maps for each cadastral parcel according to the MXD template; Specifically in this embodiment, it should be noted that the cadastral parcel attribute data packet stores cadastral parcel numbers, names of cadastral parcel right holders or cadastral parcel right holding units, cadastral parcel areas, cadastral parcel boundary line numbers, lengths of cadastral parcel boundary lines, the first set of boundary points, information about the four boundaries of the cadastral parcel, and the cadastral geographic data after identification and annotation. The data format in the data packet is in a format recognizable by ArcGIS; S7. Inspect and correct the cadastral maps: Check whether the cadastral parcel boundaries are accurate, whether the text annotations are correct, and whether the map elements are complete. If any one or more of the above are not the case, the cadastral maps need to be corrected. After the correction is completed, export the cadastral maps to a preset image format; S8. Collect data for cadastral map generation: Collect cadastral parcel ownership information correction data, cadastral map inspection and correction data, and cadastral geographic data identification data; Further, the corrected data of the land parcel ownership information includes the area deviation coefficient of each land parcel, the corrected coordinates of the first boundary point, and the true coordinates of the first boundary point; the inspection and correction data of the land parcel map includes the number of correct text annotations, the number of corrected text annotations, the number of correct map elements, the number of corrected map elements, the correct side length of the land parcel boundary, and the corrected side length of the map boundary; the identification data of the land parcel geographical data includes the number of correctly identified and annotated land parcel categories and the total number of land parcel categories.

[0027] S9. Evaluation of the quality of land parcel map generation: After processing the corrected data of the land parcel ownership information, the inspection and correction data of the land parcel map, and the identification data of the land parcel geographical data collected, evaluate the quality of land parcel map generation.

[0028] Further, the specific steps for evaluating the quality of land parcel map generation are as follows: E1. Compare the corrected coordinates Pai(xai, yai) of the i-th first boundary point with the true coordinates Pbi(xbi, ybi) of the corresponding i-th first boundary point one by one to calculate the position deviation coefficient αwi of the i-th first boundary point. The specific formula is: , and calculate the average value of the calculated position deviation coefficients to obtain the position average deviation coefficient αwe. E2. Calculate the ratio of the number of correct text annotations nax to the sum of the number of correct text annotations and the number of corrected text annotations nay, the ratio of the number of correct map elements nbx to the sum of the number of correct map elements and the number of corrected map elements nby, and the ratio of the correct side length of the land parcel boundary ncx to the sum of the correct side length of the land parcel boundary and the corrected side length of the map boundary ncy, and accumulate and calculate the average value to obtain the map production accuracy coefficient βe of the land parcel map. The specific formula is: ; E3. Calculate the ratio of the number of correctly identified and annotated land parcel categories nex to the total number of land parcel categories nz to obtain the identification accuracy coefficient γe of the land parcel geographical data. The specific formula is: ; E4. The larger the map production accuracy coefficient and the identification accuracy coefficient of the land parcel geographical data, the higher the quality of land parcel map generation. The lower the position average deviation coefficient and the area deviation coefficient, the higher the quality of land parcel map generation. Calculate the product of the map production accuracy coefficient and the identification accuracy coefficient of the land parcel geographical data, calculate the product of the compensation processing results of the position average deviation coefficient and the area deviation coefficient, and the ratio of the two is the land parcel map generation quality index Rz. The specific formula is: .

[0029] In this embodiment, it should be specifically noted that the preset values used are selected based on actual needs, and no specific value limits are made here.

[0030] Such as Figure 2This embodiment provides an AI-based automatic cadastral map generation system, including a cadastral data collection module, a cadastral ownership information correction module, a cadastral four-boundary analysis module, a cadastral geographic data recognition module, a cadastral map template creation module, a cadastral map automatic generation module, a cadastral map self-check and correction module, a cadastral map generation data collection module, a cadastral map generation quality evaluation module, and a database.

[0031] The cadastral data collection module is connected to the cadastral ownership information correction module. The cadastral ownership information correction module is connected to the cadastral four-boundary analysis module and the cadastral geographic data recognition module. The cadastral ownership information correction module, the cadastral four-boundary analysis module, the cadastral geographic data recognition module, and the cadastral map template creation module are connected to the cadastral map automatic generation module. The cadastral map automatic generation module, the cadastral map self-check and correction module, the cadastral map generation data collection module, and the cadastral map generation quality evaluation module are connected in sequence. All modules in the system are connected to the database.

[0032] The cadastral data collection module uses a radio frequency transmitter to collect cadastral ownership information and uses a drone to obtain cadastral geographic data; The cadastral ownership information correction module is used to correct the cadastral boundary line, cadastral area, and cadastral boundary points in the collected cadastral ownership information; The cadastral four-boundary analysis module uses the method of orthogonal rectangle side projection to divide the boundary line into four parts: east, south, west, and north, and obtains the cadastral four-boundary information based on the set four-boundary determination rules; The cadastral geographic data recognition module uses an AI algorithm to identify and label the collected cadastral geographic data; The cadastral map template creation module uses ArcGIS to create a cadastral map template; The cadastral map automatic generation module runs a Python script in the command prompt of ArcMap. After automatically reading the cadastral attribute data packet, the script generates a cadastral map for each cadastral plot according to the MXD template; The cadastral map self-check and correction module checks whether the cadastral boundary line is accurate, whether the text annotation is correct, and whether the map elements are complete. If any one or more of the above are not the case, the cadastral map needs to be corrected. After the correction is completed, the cadastral map is exported in a preset image format; The cadastral map generation data collection module is used to collect cadastral ownership information correction data, cadastral map inspection and correction data, and cadastral geographic data recognition data; The cadastral map generation quality evaluation module processes the collected cadastral ownership information correction data, cadastral map inspection and correction data, and cadastral geographic data recognition data, and then evaluates the quality of cadastral map generation; The database is used to store the data information of all modules in the system.

[0033] Finally, the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for automatically generating a plot map based on AI, characterized in that: The following steps are involved: S1. Collecting land parcel data: using radio frequency transmitters to collect land parcel ownership information and using drones to obtain land parcel geographic data; S2. Integrate and correct the land parcel ownership information: correct the land parcel boundary lines, land parcel areas and land parcel boundary points in the collected land parcel ownership information; S3. Analyze the four boundaries of the land parcel: Use the rectangular edge projection method to divide the boundary line into four parts: east, south, west and north, and obtain the four boundary information of the land parcel based on the set four boundary determination rules; S4. AI identifies and annotates parcel geographic data: Identifies and annotates the collected parcel geographic data based on AI algorithms; S5. Create a parcel map template: Use ArcGIS to create a parcel map template; S6. Automatic batch generation of parcel maps: Run the Python script in the ArcMap command prompt. The script automatically reads the parcel attribute data package and generates a parcel map for each parcel according to the MXD template. S7. Verification and correction of the plot map: Verify whether the plot boundary is accurate, the text annotation is correct, and the map elements are complete. If any one or more of the above situations are not met, the plot map needs to be corrected. After the correction is completed, the plot map is exported to a preset image format; S8, data collection for plot map generation: collecting plot ownership information correction data, plot map verification and correction data, and plot geographic data identification data; S9. Evaluation of the quality of plot map generation: Evaluate the quality of plot map generation after processing the collected plot ownership information correction data, plot map verification and correction data, and plot geographic data identification data.

2. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The land parcel ownership information includes the land parcel number, land parcel boundary line, first boundary point, land parcel area, land parcel owner or land parcel ownership unit name; the land parcel geographic data is the land parcel ground orthophoto real-life image data; the land parcel boundary line is the ownership boundary line between the land parcel and the adjacent land parcel.

3. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The specific steps of collecting parcel data are as follows: A1. Issue a number of radio frequency transmitters to land parcel owners or persons in charge of land parcel ownership units. The radio frequency transmitters have the functions of transmitting radio frequency communication signals and inputting information; A2. After receiving the radio frequency transmitter, the land parcel owner or the person in charge of the land parcel ownership unit enters the land parcel number, the name of the land parcel owner or the land parcel ownership unit, the land parcel area and the radio frequency transmitter number information in the radio frequency transmitter, and arranges the radio frequency transmitters at the turning points of the land parcel boundary line and the intersections with the adjacent land parcel boundary lines in the order of numbers. The radio frequency transmitters arranged at the turning points of the land parcel boundary line are numbered a1, a2, ..., an, and the radio frequency transmitters arranged at the intersections with the adjacent land parcel boundary lines are numbered b1, b2, ..., bn. The radio frequency transmitters at both the turning points of the land parcel boundary line and the intersections with the adjacent land parcel boundary lines have two numbers at the same time; A3. The input information of the radio frequency transmitter is transmitted as a radio frequency communication signal, and the drone is controlled to use the configured high-definition camera equipment according to the set flight route to obtain the orthophoto real-life image data of the land parcel area. During this period, the radio frequency communication signal receiving device arranged on the drone receives the radio frequency communication signal sent by each radio frequency transmitter, and obtains the land parcel number, the name of the land parcel owner or the land parcel ownership unit, the land parcel area, the radio frequency transmitter number information and the geographical location coordinates of the radio frequency transmitter input in the received radio frequency transmitter; A4. Connect the corresponding geographical location coordinates in sequence according to the numbers of the radio frequency transmitters arranged at the turning points of the land boundary lines. The closed lines generated are the land boundary lines. The boundary lines of segments a1a2, a2a3, ..., and ana1 are numbered x1, x2, ..., xn, respectively. The geographical location coordinate points and numbers of the radio frequency transmitters arranged at the turning points of the land boundary lines are marked as the third boundary points and corresponding numbers. The geographical location coordinate points and numbers of the radio frequency transmitters arranged at the intersections with the adjacent land boundary lines are marked as the first boundary points and corresponding numbers.

4. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The specific steps of integrating and correcting the parcel ownership information are as follows: B1. Retrieve the parcel boundary lines and the adjacent parcel boundary lines, and observe whether there are other overlapping areas between the corresponding included parcel area and the adjacent parcel area except the boundary lines. If there are, the parcel boundary line correction task is triggered. If not, the parcel boundary line correction task is not triggered, and the parcel area correction link is directly entered; B2. When entering the land parcel area correction phase, check the coordinates and corresponding numbers of each layout point of the RF transmitter to ensure that the coordinates and numbers of the RF transmitter are arranged in a clockwise or counterclockwise direction according to the boundary line. Then substitute the coordinates of the RF transmitter into the Gaussian area formula to calculate the land parcel area Ax. Calculate the area deviation coefficient αm based on the calculated land parcel area and the received land parcel area. The area deviation coefficient is the ratio between the absolute value of the difference between the calculated land parcel area Ax and the received land parcel area As and the smaller value of the two. The specific formula is: When the calculated plot area Ax is the same as the received plot area As, αm=0, and the calculated area deviation coefficient is compared with the preset area deviation upper limit. If the calculated area deviation coefficient is greater than the area deviation upper limit, the plot area correction task is triggered. Otherwise, the plot area correction task is not triggered, and the plot boundary point correction phase is entered; B3, extracting the first boundary point to constitute a point set, taking the first boundary point numbered b1 as the starting point, calculating the boundary point straight line distance D1 between the first boundary point numbered b1 and the first boundary point numbered b2 based on the point coordinates and the Euclidean distance formula, and obtaining the boundary line distance L1 between b1 and b2, calculating L1 / D1 and comparing the calculation result with a preset value, if the calculated value is greater than the preset value, it is determined that a new boundary point needs to be inserted between b1 and b2, otherwise it is not necessary to insert a new boundary point, and the new boundary point insertion determination between the first boundary point numbered b2 and the first boundary point numbered b3 is continued; B4. When it is determined that a new boundary point needs to be inserted between b1 and b2, the point with the farthest distance from the boundary line of b1 and b2 to the straight line connecting b1 and b2 is found and inserted as the new boundary point between b1 and b2, and the new boundary point is marked as the first boundary point and numbered as b2. The first boundary points originally numbered b2, b3, ..., bn are automatically numbered backward by one position and changed to b3, b4, ..., bn+1; B5. Repeat steps B3 and B4 until the ratio of the boundary line distance to the straight line distance of any adjacent first boundary points is less than or equal to the preset value, and the point set correction of the first boundary point is completed.

5. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The specific steps of analyzing the boundaries of the plot are as follows: C1. Use the rectangular edge projection method to first divide the boundary line into four parts: east, south, west and north; C2. For all neighboring plots of the present plot, if a neighboring plot has a common edge with the present plot in any direction of the southeast, northwest, or northeast of the present plot, then the neighboring plot will be included in the boundary information of the corresponding direction of the present plot; if a neighboring plot has common edges with multiple parts of the present plot at the same time, then the neighboring plot will be included in the boundary information of the present plot in multiple directions.

6. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The specific operations of the AI ​​in identifying and labeling the geographic data of the plot are as follows: identifying the physical elements within the plot, highlighting the edge contours of the buildings, structures, roads and terrain features within the plot, highlighting the boundary lines and boundary points, marking the buildings within the plot with the building names as data labels, marking the structures within the plot with the structure names as data labels, marking the terrain features within the plot with the terrain feature names as data labels, marking each boundary line with the boundary line number and boundary line side length as data labels, and marking the boundary points with the boundary point number and coordinates.

7. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The parcel ownership information correction data includes the area deviation coefficient of each parcel, the corrected coordinates of the first boundary point and the true coordinates of the first boundary point; the parcel map verification correction data includes the correct number of text annotations, the correct number of text annotations, the correct number of map elements, the correct number of map elements, the correct length of the parcel boundary line and the correct length of the map boundary line; the parcel geographic data identification data includes the number of parcel object types correctly identified and marked and the total number of parcel object types.

8. The method for automatically generating a plot map based on AI according to claim 1, characterized in that: The specific steps of the plot map generation quality evaluation are as follows: E1. Compare the corrected coordinates Pai (xai, yai) of the i-th first boundary point with the real coordinates Pbi (xbi, ybi) of the corresponding i-th first boundary point one by one to calculate the position deviation coefficient αwi of the i-th first boundary point. The specific formula is: , calculate the average value of the calculated position deviation coefficients to obtain the position average deviation coefficient αwe; E2. Calculate the ratio of the number of correct text annotations nax to the sum of the number of correct text annotations and the number of correct text annotations nay, the ratio of the number of correct map elements nbx to the sum of the number of correct map elements and the number of correct map elements nby, the ratio of the correct parcel boundary side length ncx to the sum of the correct parcel boundary side length and the correct map boundary side length ncy, and add them up to get the average value to obtain the parcel map production accuracy coefficient βe. The specific formula is: ; E3. Calculate the ratio of the number of correctly identified and marked land parcel object types nex to the total number of land parcel object types nz to obtain the land parcel geographic data identification and marking accuracy coefficient γe. The specific formula is: ; E4. The greater the accuracy coefficient of plot map production and the accuracy coefficient of plot geographic data identification and annotation, the higher the quality of plot map generation. The lower the average position deviation coefficient and area deviation coefficient, the higher the quality of plot map generation. Calculate the product of the accuracy coefficient of plot map production and the accuracy coefficient of plot geographic data identification and annotation, and calculate the product of the compensation processing result of the average position deviation coefficient and the compensation processing result of the area deviation coefficient. The ratio of the two is the plot map generation quality index Rz. The specific formula is: .

9. An AI-based automatic plot map generation system, implementing an AI-based automatic plot map generation method according to any one of claims 1 to 8, characterized in that: include: Parcel data collection module: Use radio frequency transmitters to collect parcel ownership information and use drones to obtain parcel geographic data; Land parcel ownership information correction module: used to correct the land parcel boundary lines, land parcel areas and land parcel boundary points in the collected land parcel ownership information; Parcel boundary analysis module: Use the rectangular edge projection method to divide the boundary line into four parts: east, south, west and north, and obtain the parcel boundary information based on the set boundary determination rules; Parcel geographic data identification module: Identify and annotate the collected parcel geographic data based on AI algorithm; Parcel map template creation module: Use ArcGIS to create parcel map templates; Parcel map automatic generation module: Run the Python script in the ArcMap command prompt. The script automatically reads the parcel attribute data package and generates a parcel map for each parcel according to the MXD template. Parcel map self-checking and correction module: Check whether the parcel boundary is accurate, whether the text annotation is correct, and whether the map elements are complete. If any one or more of the above situations are not met, the parcel map needs to be corrected. After the correction is completed, the parcel map will be exported to a preset image format; Parcel map generation data collection module: used to collect parcel ownership information correction data, parcel map verification and correction data, and parcel geographic data identification data; Plot map generation quality evaluation module: evaluates the quality of plot map generation after processing the collected plot ownership information correction data, plot map verification and correction data, and plot geographic data identification data.

Images